摘要:
An embodiment of the invention is a method for proactive failover using user-defined rules. An event log of a first server node is monitored to check for user-specified application events. One of the user-specified application events corresponding to an impending failure in an application running on a first server node is detected. In automatic response to the detected impending failure, a proactive failover process is executed to transfer the application to a second server node for continued execution, the second server node being connected to the first server node in a cluster.